Elucidating Latent Mechanistic Complexity in Competing Acid-Catalyzed Reactions of Salicylaldehyde-Derived Baylis-Hillman Adducts

Article abstract of DOI:10.1021/acs.joc.5b02372

¹H NMR-based kinetic studies have revealed the latent mechanistic complexity of deceptively simple hydrochloric acid-catalyzed reactions of salicylaldehyde-derived Baylis-Hillman adducts. Reactions conducted at 0 °C afforded 2-(chloromethyl)cin

Full text of DOI:10.1021/acs.joc.5b02372

Article
pubs.acs.org/joc
Elucidating Latent Mechanistic Complexity in Competing Acid-
Catalyzed Reactions of Salicylaldehyde-Derived Baylis−Hillman
Adducts
Temitope O. Olomola,† Rosalyn Klein,† Mino R. Caira,‡ and Perry T. Kaye*,†
†Department of Chemistry and Centre for Chemico- and Biomedicinal Research, Rhodes University, Grahamstown 6140, South
Africa
‡Department of Chemistry, University of Cape Town, Rondebosch 7701, South Africa
S
* Supporting Information
1
ABSTRACT: H NMR-based kinetic studies have revealed the latent
mechanistic complexity of deceptively simple hydrochloric acid-
catalyzed reactions of salicylaldehyde-derived Baylis−Hillman adducts.
Reactions conducted at 0 °C afforded 2-(chloromethyl)cinnamic acid
derivatives as the major products and the corresponding 3-
(chloromethyl)coumarin derivatives as the minor products. In reactions
conducted in refluxing acetic acid, however, the 3-(chloromethyl)-
1
coumarin derivatives are the sole products. Variable-temperature H
NMR analysis permitted the determination of the rate constants and
kinetic parameters involved in the pseudo-first-order formation of (Z)-
2-(chloromethyl)-3-(2-hydroxyphenyl)-2-propenoic acid. The kinetic
data clearly preclude the operation of classical kinetic versus
thermodynamic control and indicate the operation of three independent
reaction pathways. Theoretical studies of these pathways undertaken at
the B3LYP/6-31G(d) level permitted rationalization of the experimental data and provided insights into the possible mechanism
of the enzymic E−Z isomerization and cyclization of (E)-cinnamic acid analogues to afford coumarins.
INTRODUCTION
■
Numerous applications of Baylis−Hillman1 (or Morita−Baylis−
Hillman2) methodology continue to be published, and the
reaction has been the subject of regular reviews.3 Our
contributions in this area have focused on the synthesis of
benzannulated heterocycles, including quinolines,4 coumarins,5
and chromenes,6 and their structural elaboration to compounds
with medicinal potential. We have recently reported the
synthesis7 and preliminary evaluation8 of Baylis−Hillman-
afforded α-chloromethyl-α,β-unsaturated products, apparently
derived coumarin−AZT conjugates, such as compound 1, as
potential dual-action HIV-1 protease/reverse transcriptase (PR/
RT) inhibitors. Structural similarities between the HIV-1
integrase (IN) and RNase H active sites9 have prompted
research on the development of dual-action RT/IN inhibitors,
and Wang et al.10 reported the synthesis of the HIV-1 RT/IN
inhibitor 2, which exhibited encouraging inhibitory activities
(RT, IC50 = 57 nM; IN, IC50 = 2.4 μM). The apparent HIV-1 IN
inhibition activity of L-chicoric acid (3) and caffeic acid phenethyl
ester has been attributed to the presence of the cinnamate ester
moieties,11 and we have been examining the use of Baylis−
Hillman adducts to access cinnamate ester−AZT conjugates,
such as compounds 8 (Scheme 1), as potential dual-action HIV-1
RT/IN inhibitors.
via a conjugate addition−elimination pathway, and it seemed
reasonable to explore this approach to access the desired
cinnamate ester−AZT conjugates 8a−e (Scheme 1). The
concomitant formation of both 2-(chloromethyl)cinnamic
acids 6a−e and 3-(chloromethyl)coumarin derivatives 7a−e
and the ready cyclization of compound 6a to give coumarin
derivative 7a at elevated temperature prompted us to undertake
detailed kinetic and theoretical studies to elucidate the
mechanistic details. While some attention has been given to
studying the mechanism of the Baylis−Hillman reaction per se,13
relatively little appears to have been given to transformations of
the resulting versatile multifunctional adducts. In this paper, we
discuss the results of our experimental kinetic and theoretical
We have previously observed12 that treatment of methyl vinyl
ketone (MVK)-derived Baylis−Hillman adducts with HCl
Received: October 15, 2015
Published: December 14, 2015
© 2015 American Chemical Society
109
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
Scheme 1
Figure 1. Crystal structure of cinnamic acid derivative 6a, showing the crystallographic numbering and the crystal packing arrangement. Thermal
displacement ellipsoids are drawn at the 50% probability level.
studies of the formation of compounds 6a and 7a from the
Baylis−Hillman adduct 5a.
O11i (i = 1 − x, −y, −z) and O12−H12···O11ii (ii = −x, 1 − y,
−z) with O···O distances of 2.901(4) and 2.579(4) Å,
respectively. These create infinite ribbons (portions of two
ribbons are shown in Figure 1) that contain alternating
centrosymmetric R22(8) H-bonded motifs (the carboxylic acid
dimers)15 and R22(16) motifs.16 A layered crystal structure results,
with only van der Waals forces providing cohesion between
successive layers.
After the structures of the major and minor products 6a−e and
7a−e, respectively, were ascertained, attention was given to
exploring the mechanistic pathways involved in the formation of
these products and the reasons for the observed chemo-
selectivity. The reaction of the parent system 5a with HCl in
methanol-d4 was monitored by 1H NMR spectroscopy using the
signals summarized in Table 1 as probes for the substrate 5a, the
products 6a and 7a, and the putative intermediates 10a and 13a.
RESULTS AND DISCUSSION
■
Unlike their methyl and ethyl ester analogues, the tert-butyl
acrylate-derived Baylis−Hillman adducts 5a−e are isolable,
permitting controlled subsequent elaboration,14 and they were
prepared from the corresponding 2-hydroxybenzaldehydes 4a−e
as described previously.14 The resulting adducts 5a−e were
treated with HCl [generated in situ by adding acetyl chloride
CAUTIOUSLY (!) to dry ethanol at 0 °C; Scheme 1]. In each
case, two products were isolated in excellent overall yield (95−
100%): a 2-(chloromethyl)cinnamic acid derivative 6 as the
major product and the corresponding 3-(chloromethyl)-
coumarin 7. Interestingly, refluxing the crude mixture of
compounds 6a and 7a in acetic acid for 1 h resulted in complete
conversion to the coumarin derivative 7a. In all cases, the
assigned structures are supported by the HRMS, elemental, and
NMR spectroscopic data. Compounds 6a and 7a were separated
by fractional recrystallization, and the structure of compound 6a
was confirmed unequivocally by single-crystal X-ray analysis,
which indicated a Z configuration about the double bond (Figure
1).
Table 1. Origins and Ranges of 1H NMR Chemical Shifts Used
as Probes To Determine the Concentrations of the Substrate,
Intermediates, and Products in Methanol-d4
compound
signal
δ range (ppm)
5a
3′-methylene
ArH
5.797−5.866
7.933−8.029
4.565−4.513
8.286−8.357
5.585−5.667
10a
6a
Salient molecular parameters of 6a include the length of the
C8−C9 double bond [1.347(5) Å] and the C3−C2−C8−C9
dihedral angle [29.8(6)°] between the two planar residues. In the
crystal, the unique intermolecular hydrogen bonds are O7−H7···
3′-methylene
ArH
7a
13a
4-H
110
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
Figure 2. Partial time-dependent 600 MHz 1H NMR spectra selected at 300 s intervals from the commencement of the reaction of the Baylis−Hillman
adduct 5a with in situ-generated HCl in methanol-d4 at 276.2 K, illustrating some of the signals assigned to structures 5a, 6a, 7a, 10a, and 13a.
Figure 3. Logarithmic pseudo-first-order kinetic plots reflecting the consumption of the substrate 5a at different temperatures (276.2−293.4 K); [HCl]
= 2.35 M.
A partial stack plot reflecting changes in the concentrations of the
substrate 5a and the major product 6a is shown in Figure 2.
Analysis of the H NMR data reveals that consumption of
ΔS⧧; Table 4]. Similar treatment of the data for the pseudo-first-
order consumption of intermediate 13a (leading to the minor
product 7a) provided the corresponding rate constants at
different temperatures (Table 3), and Arrhenius (R2 = 0.9436;
Figure 6) and Eyring (R2 = 0.9402) plots afforded the kinetic
parameters (Table 4).
While the rate constants (kobs) for the consumption of
compound 10a at different temperatures are ca. 6 times smaller
than those for compound 13a (Tables 2 and 3), the
corresponding Eyring plots and Arrhenius plots (Figures 5 and
6) exhibit almost identical slopes. As a result, the Ea and ΔH⧧
1
substrate 5a follows pseudo-first-order kinetics over a range of
temperatures (276.2−293.4 K; Figure 3). It is also apparent that
the transformation of intermediate 10a to compound 6a from
9000 s onward follows pseudo-first-order kinetics (Figure 4).
Arrhenius (R2 = 0.8848; Figure 5) and Eyring (R2 = 0.8783) plots
of the rate constants for the consumption of intermediate 10a
obtained at different temperatures (276.2−293.4 K; Table 2)
afforded the kinetic parameters [Ea, ΔG⧧ (at 298 K), ΔH⧧, and
111
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
On the other hand, cyclization via path B (described below) is
proposed to involve O-acyl cleavage, i.e., acyl substitution to
afford the cyclized intermediate 12a followed by dehydration to
afford coumarin derivative 13a. While the rate of formation of the
major product, 2-(chloromethyl)cinnamic acid 6a, clearly
depends on the concentration of cinnamic acid intermediate
10a, the rate of formation of the minor product, 3-
(chloromethyl)coumarin 7a, depends on the concentration of
cyclic intermediate 13a, which reaches a maximum before
intermediate 10a. Once intermediate 13a is consumed, further
production of the minor product 7a ceases. We have shown,
however, that the transformation 6a → 7a can be effected at
elevated temperature in refluxing acetic acid. Consequently, on
the basis of a combination of a priori considerations and the
Figure 4. Logarithmic pseudo-first-order kinetic plot from 9000 s
onward reflecting the consumption of intermediate 10a leading to the
major product 6a at 286.45 K; [HCl] = 2.35 M.
1
consolidated H NMR kinetic data illustrated in Figure 7, the
transformations are considered to involve ester cleavage and
conjugate addition−elimination (or an SN′ reaction) (path A;
Scheme 2); cyclization−dehydration, i.e., lactonization (path B);
and isomerization−cyclization (path C).
In order to elucidate the mechanistic implications of these
pathways more fully, theoretical studies at the B3LYP/6-31G(d)
level were undertaken using Gaussian 03.17 In all, over 20
stationary-state structures were located, geometry-optimized,
and characterized as substrates, products, or intermediates (with
no imaginary frequencies) and transition-state complexes (with a
single imaginary frequency). The results of these computational
studies are summarized in Table 5 and illustrated graphically in
Figures 8−11.
Path A: Formation of the Major Product 6a via O-Alkyl
Cleavage. A relatively low energy transition-state complex (TS
I) was located for the initial HCl-catalyzed O-alkyl cleavage of the
substrate 5a, consistent with the rapid formation of β-hydroxy
acid intermediate 10a (Figure 8). The transition-state complex is
stabilized by extended hydrogen bonding between the phenolic
and alcohol hydroxyl groups and the carboxylic acid carbonyl
oxygen. The chloride anion is not only hydrogen-bonded to the
incipient carboxylic acid proton (O−H···Cl = 1.77 Å) but also
spatially poised (2.25 Å away) to remove a proton from the
essentially trigonal-planar tert-butyl cation and thus release
isobutylene. Not surprisingly, the transition-state complex TS I
collapses to a stable arrangement of the three discrete species,
viz., intermediate 10a, isobutylene, and HCl, as depicted in
Figure 8. Conjugate addition of HCl to intermediate 10a leads to
intermediate enediol 11a via transition-state complex TS II, in
which the reacting moieties are engaged in a delocalized six-
center association. The kinetic parameters calculated using the
stable three-component “complex” of 10a, isobutylene, and HCl
as the substrate afford a free energy of activation that is very close
to the experimentally determined value for the consumption of
10a and a positive entropy of activation (Table 4), consistent
with loss of isobutylene from the “complex”.
Figure 5. Arrhenius plot (ln k against 1/T) for the consumption of
intermediate 10a leading to the major product 6a.
values for the two reactions correspond almost exactly to two
decimal places (Table 4)!
In competing reactions, the observation of a temperature-
dependent change in the identity of the major product often
suggests the operation of classical kinetic versus thermodynamic
control. In this case, however, careful examination of the
consolidated kinetic data (Figure 7) challenges such an
assumption on several grounds, viz., (i) the apparent absence
of a common intermediate; (ii) the absence of any evidence of
reversibility (5a ⇄ 6a ⇄ 7a) under the reaction conditions; and
(iii) the relative rates and sequence of formation of the products
6a and 7a. The fact that formation of the minor product 7a
decreases rapidly and then ceases altogether following
consumption of the substrate 5a indicates that it is produced
independently of the major product 6a. In the absence of water
(in the early stages of the reaction, at least), cleavage of the ester
moiety in protonated ester 9a is expected to involve O-alkyl
cleavage to afford carboxylic acid intermediate 10a (Scheme 2).
Table 2. Rate Constant Data (kobs) at Various Temperatures Used for the Consumption of 10a
a
T (K)
kobs (s−1
)
R2
k (M−1 s−1
)
1/T (K−1
)
ln k
ln(k/T)
276.21
280.05
282.61
286.45
289.00
293.40
0.000017
0.000023
0.000059
0.000077
0.000102
0.000107
0.999162
0.999498
0.999766
0.999916
0.999879
0.999459
7.23401 × 10−6
9.78719 × 10−6
2.51063 × 10−5
3.27658 × 10−5
4.34041 × 10−5
4.55317 × 10−5
0.003620434
0.003570791
0.003538445
0.003491011
0.003460208
0.003408316
−11.83671704
−11.53443617
−10.59239294
−10.32612496
−10.04495757
−9.997101552
−17.45787849
−17.16940433
−16.2364608
−15.98368897
−15.71138426
−15.67863842
a
k = kobs/[HCl]; [HCl] = 2.35 M.
112
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
Table 3. Rate Constant Data (kobs) at Various Temperatures Used for the Consumption of 13a
a
T (K)
kobs (s−1
)
R2
k (M−1 s−1
)
1/T (K−1
)
ln k
ln(k/T)
276.21
280.05
282.61
286.45
289.00
293.40
0.000099
0.000122
0.000254
0.000394
0.000504
0.000621
0.999382
0.999674
0.993677
0.997412
0.997882
0.993978
4.21275E-05
5.19147E-05
0.000108085
0.000167659
0.000214467
0.000264254
0.003620434
0.003570791
0.003538445
0.003491011
0.003460208
0.003408316
−10.07481054
−9.865909341
−9.132596119
−8.693579477
−8.447354118
−8.238599304
−15.69597198
−15.5008775
−14.77666397
−14.35114348
−14.11378081
−13.92013617
a
k = kobs/[HCl]; [HCl] = 2.35 M.
Hillman adduct 5a to tetrahedral intermediate 12a involves
nucleophilic attack of the phenolic oxygen on the ester acyl
center via transition-state complex TS IV (Figure 9). In this
process, HCl catalyzes the transformation by simultaneously (i)
protonating the ester carbonyl oxygen, thus enhancing the
electrophilicity of the acyl carbon; (ii) deprotonating the
phenolic group, thus enhancing the nucleophilicity of the
phenolic oxygen; and (iii) facilitating formation of a six-center
hydrogen-bonded chelate in which the reactive centers are ideally
disposed relative to each other at a distance of 1.78 Å, thus
decreasing the entropic demand. The next step, which proceeds
via transition-state complex TS V, involves acid-catalyzed loss of
tert-butyl alcohol from intermediate 12a and formation of 4-
hydroxy-3-methylenepyran-2-one 13a. The structural changes
associated with the imaginary frequency (363.2i cm−1) for
transition-state complex TS VI in the final step (13a → 7a)
exemplify a classic acid-catalyzed SN2′ reaction involving (i)
protonation of the 3-hydroxyl oxygen, (ii) nucleophilic attack by
Cl− at the terminal vinylic center, (iii) migration of the double
bond, and (iv) loss of the syn leaving group as neutral H2O. The
minor product, 3-chloromethylcoumarin 7a, is thus obtained in
what is clearly an overall exergonic process (ΔG° = −28.18 kcal
mol−1). The calculated and experimental free energies of
activation for the consumption of intermediate 13a correlate
closely (Table 4); the ΔS⧧ values are both negative, but the
theoretical model underestimates the Ea and ΔH⧧ values.
Comparative free energy profiles for paths A and B are
presented on the same scale in Figure 10. It is immediately
apparent that the theoretical data seem to indicate that the major
product 6a is neither thermodynamically nor kinetically favored!
In fact, the major product 6a is calculated to be significantly less
stable (by ΔG° = 42.8 kcal mol−1) than the minor product 7a,
and the transition-state complexes leading to the major product
(6a; path A) have consistently higher free energies than those
leading to the minor product (7a; path B). However, the
apparent anomalies presented by this comparison of the
theoretical results with the experimental data can be explained.
First, it may be argued that following the rapid formation of
intermediate carboxylic acid 10a and the concomitant irrever-
sible loss of isobutylene in the first step in path A, consumption of
Figure 6. Arrhenius plot (ln k against 1/T) for the consumption of
intermediate 13a leading to the minor product 7a.
Dehydration of enediol 11a via transition-state complex TS III
finally affords the major product 6a. Transition-state complex TS
III, which corresponds to the highest point of the free energy
profile, is stabilized by an interesting hydrogen-bonding
networka level of molecular organization that would account
for the experimentally observed negative entropy of activation for
the overall transformation 10a → 6a (Table 4). We thus conclude
that rapid and irreversible O-alkyl cleavage of the substrate 5a
affords the stable β-hydroxy acid intermediate 10a, which is
slowly consumed in the reaction sequence 10a → 11a → 6a, a
pattern consistent with the experimental results illustrated in
Figure 7. [The preference for a conjugate addition−elimination
sequence rather than SN′ displacement of the protonated
hydroxyl group by the chloride anion is supported by
comparative studies on related systems12 and the relative
condensed local softness values for the carbonyl (3.97) and
hydroxylic (0.91) carbons using the electrophilic Fukui functions
(f+)18 evaluated at a B3LYP/6-31G(d) level.] Noteworthy in the
formation of the major isomer 6a is the multifaceted role of HCl
in (i) acid catalysis of O-alkyl cleavage of tert-butyl ester 5a, (ii)
deprotonation (by the conjugate base Cl−) of the tert-butyl
cation, (iii) conjugate addition to α,β-unsaturated acid 10a, and
(iv) acid catalysis of the dehydration of enediol 11a.
Path B: Formation of the Minor Product 7a via O-Acyl
Cleavage. Reversible HCl-catalyzed cyclization of Baylis−
Table 4. Kinetic Parameters Obtained from Arrhenius and Eyring Plots of the Data for Consumption of Intermediates 10a and 13a
consumption of 10a
consumption of 13a
a
bc
,
a
b
experimental
theoretical
experimental
theoretical
b
b
ΔG⧧ (kcal mol−1
ΔS⧧ (cal mol−1 K−1
Ea (kcal mol−1
ΔH⧧ (kcal mol−1
)
22.89 0.31
23.65
20.83
29.27
29.86
21.91 0.20
−11.84 4.10
18.95 0.20
18.39 0.20
21.04
−35.75
10.97
)
−15.09 6.1
)
18.96 0.30
)
18.39 0.30
10.38
a
b
c
Errors were calculated from the respective standard errors for the slope and intercept for each graph. At 298.15 K. Results are based on the use of
the stable three-component complex of 10a, isobutylene, and HCl.
113
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
Figure 7. Plots of concentrations against time (from 600 MHz 1H NMR spectra in methanol-d4) for the reaction of the Baylis−Hillman adduct 5a with
in situ-generated HCl at 276.2 K, showing the consumption of the substrate and the formation of intermediates 10a and 13a, the major product 6a, and
the minor product 7a.
Scheme 2. Proposed Reaction Pathways for the Formation of Compounds 6a and 7a from the Baylis−Hillman Adduct 5a in HCl/
EtOH and the Formation of 7a from 6a in Refluxing AcOH
intermediate 10a leads slowly to the product 6a; the rate of
irreversible consumption of the substrate 5a in this reaction path
may be expressed by eq 1. Second, the rate-determining step in
path B (i.e., the second step via transition-state complex TS V) is
preceded by what is expected to be an equilibrium between the
substrate 5a and the cyclized tetrahedral intermediate 12a; the
rate of this reaction is therefore proportional to the product of
the equilibrium constant KB1 and the rate constant kB2 (eq 2).
Under the assumption of pseudo-first-order reactions (loss of
HCl is considered to be minimal), the free energy of activation
for the first step in path A (ΔG⧧ = 24.55 kcal mol−1) corresponds
to a rate constant kA1 = 6.32 × 10−6 s−1. In path B, the free energy
difference between 5a and 12a (ΔG° = 7.87 kcal mol−1 at 298.15
K) corresponds to an equilibrium constant KB1 = 1.69 × 10−6,
while the free energy of activation for the second step in path B
the substrate 5a should be consumed more rapidly (ca. 12-fold)
in the irreversible formation of intermediate 10a in path A than in
the formation of intermediate 13a in the rate-determining
second step in path Ba result that corresponds reasonably
closely to the experimentally determined ratio of the two products
(7a:6a = 1.0:0.11) at the end of the kinetic run at 293 K (at which
stage, a small concentration of 10a still remains!). Thus,
formation of the minor product 7a ceases early in the reaction,
while formation of the major product 6a continues slowly until
intermediate 10a is consumed.
path A (5a + HCl → 10a):
reaction rate = kA1[5a][HCl]
(1)
path B (5a + HCl → 13a):
(ΔG⧧ = 18.17 kcal mol−1) corresponds to a rate constant kB2
3.00 × 10−1 s−1; thus, KB1 × kB2 = 5.08 × 10−7 s−1. Consequently,
=
reaction rate = KB1 × kB2[5a][HCl]
(2)
114
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
Table 5. B3LYP/6-31G(d) Stationary-State Free Energies and Transition-State Vibrational Frequencies for the Geometry-
Optimized Structures Used, after Mass Balancing, in Figures 8−11
free energy (G)
structure
hartrees
kcal mol−1
frequency (cm−1
)
Path A (Figure 8)
5a
−844.942701
−1305.710485
−1305.730337
−1148.575447
−1148.58451
−1148.569654
−1072.203627
−157.117209
−76.405453
−530209.5718
−819345.7336
−819358.1909
−720742.0045
−720747.6916
−720738.3693
−672817.9619
−98592.54126
−47945.14761
−289160.708
−
TS I
−148.59
10a + HCl + isobutylene
−
TS II
11a
−148.10
−
TS III
6a
−902.51
−
−
−
isobutylene
H2O
HCl
−460.806901
Path B (Figure 9)
5a
−844.942701
−1305.713788
−1305.737054
−1305.708105
−611.400499
−1072.173867
−995.82541
−530209.5718
−819347.8063
−819362.4059
−819344.2401
−383659.6214
−672799.2872
−624889.9051
−146563.4023
−289160.708
−
TS IV
12a
−388.58
−
TS V
13a
−174.60
−
−363.18
−
TS VI
7a
t-BuOH
HCl
−233.563639
−460.806901
−
Path C (Figure 11)
6a-AcOH
TS VII
14a
−1301.249682
−1301.223523
−1301.221498
−1301.195513
−1301.244558
−1301.237301
−995.82541
−816546.5373
−816530.1223
−816528.8516
−816512.5458
−816543.322
−816538.7681
−624889.9051
−143729.1879
−
−739.44
−
TS VIII
15a
−111.23
−
−716.67
−
TS IX
7a
AcOH
−229.047031
−
Path C: Z−E Isomerization of 2-(Chloromethyl)-
cinnamic Acid Derivative 6a and Subsequent Cyclization
To Give 3-(Chloromethyl)coumarin 7a. Transformation of
(Z)-2-(chloromethyl)cinnamic acid derivative 6a to 3-
(chloromethyl)coumarin 7a requires a change in the double-
bond geometry from Z to E prior to cyclization. Although it does
not occur at low temperature (Figure 7), the transformation 6a
→ 7a is readily effected in refluxing acetic acid (bp 117 °C). It is
not unreasonable to assume that under these conditions
substrate−solvent interactions facilitate both the isomerization
and lactonization steps, and the initial stationary-state structure
6a-AcOH (Figure 11) reflects chelation of an acetic acid
molecule with 6a; an intramoleclar hydrogen-bonding inter-
action is also evident between chlorine and the cinnamic acid
proton. For practical purposes, the hydrogen-bonded chelate 6a-
AcOH was treated as the substrate in this reaction. The acetic
acid plays a pivotal role as an amphoteric catalyst in the formation
of metastable quinone methide intermediate 14a, via transition-
state complex TS VII, simultaneously protonating the cinnamic
acid carbonyl oxygen and deprotonating the phenolic group.
[The free energy of the metastable species 14a, which has no
imaginary frequencies, is slightly higher (by ca. 1 kcal mol−1) than
that of the preceding transition-state complex TS VII, but its
electronic energy is slightly lower (also by ca. 1 kcal mol−1).] The
π-electron delocalization leading to the quinone methide system
14a affords a rotatable σ bond that permits Z → E isomerization
via transition-state complex TS VIII, which is characterized by a
torsion angle of ca. 90° between the rotating planes. The
intermediate structure 15a was located, in which chelation of the
resulting (E)-cinnamic acid derivative involves both the
hydroxylic and carbonyl oxygens of the acetic acid; intra-
molecular hydrogen-bonding between chlorine and the cinnamic
acid proton is also retained. The reorientation of the chelating
moiety in structure 15a provides a level of preorganization that
facilitates cyclization and serves to stabilize the subsequent
transition-state complex TS IX via an interesting hydrogen-
bonding network. In fact, the transition-state complex TS IX is
characterized by the involvement of no less than eight centers,
with acetic acid once more serving as an amphoteric catalyst
enhancing both the electrophilicity of the acyl carbon by
protonating the acyl oxygen and the nucleophilicity of the
phenolic oxygen by removing the phenolic proton (Figure 12a).
E−Z isomerization of cinnamic acid analogues is involved in
various established syntheses of coumarin derivatives, including
the Perkin,19 Knoevenagel,20 and Wittig21 reactions, and Jeon et
al.22 recently reported the preparation of 3,4-dihydrocoumarins
via the p-toluenesulfonic acid-catalyzed thermal isomerization
and cyclization of cinnamic acid derivatives in sealed tubes at 140
°C. E−Z isomerization is also involved in the biosynthesis of
coumarin analogues from o-coumaric acid.23 In early studies,
Haskins et al.24 reported that the E−Z isomerization of β-D-
glucosyl o-hydroxycinnamic acid in sweetclover (Melilotus alba)
leaves was photochemically, not enzymically, induced. In a more
recent study, however, Bayoumi et al.25 provided a review of
115
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
Figure 8. Mass-balanced free energy profile for the formation of the major product 6a via O-alkyl cleavage (path A), showing ball-and-stick structures for
the transition states TS I−TS III and the stable arrangement of the three discrete species 10a, isobutylene, and HCl, colored according to atom type (H,
white; C, gray; O, red; Cl, green).
differing opinions concerning enzymic or photochemical
involvement in the E−Z isomerization stage of the biosynthesis
of coumarin analogues and clearly demonstrated that E−Z
isomerization of isotopically substituted (E)-cinnamic acid in
cassava (Manihot esculenta Crantz) roots is enzymatically, not
photochemically, induced! Such isomerizations are attributed to
addition−σ-rotation−elimination sequences. Somewhat surpris-
ingly, no attention appears to have been given to the possibility of
an enzymatically generated quinone methide intermediate (e.g.,
structure 16 in Figure 12b) in which appropriately placed amino
acid residues or water molecules in an enzyme active site could
provide the hydrogen-bonding environments needed to mediate
both E−Z isomerization and subsequent cyclization via the
proximate o-phenolic oxygen and the acyl center (as in structure
17)clearly entropically advantageous arrangements.
observation of potential significance in the biosynthesis of
coumarin analogues from o-coumaric acid.
EXPERIMENTAL SECTION
■
General Information. NMR spectra were recorded at 303 K
(except where described otherwise for kinetics experiments) in DMSO-
d6, CDCl3, or methanol-d4 and calibrated using solvent signals [δH: 7.26
ppm for residual CHCl3, 2.50 ppm for residual DMSO, and 3.31 ppm for
residual methanol; δC: 77.0 ppm for CDCl3 and 39.5 for DMSO-d6];
coupling constants are given in hertz. Melting points were measured
using a hot-stage apparatus and are uncorrected. IR spectra were
recorded with a diamond window, and compounds were analyzed neat.
The known tert-butyl 3-hydroxy-3-(2-hydroxyphenyl)-2-methylene-
propanoate esters 5a−e14 were prepared using both established and
microwave-assisted methods; the latter afforded the products much
more rapidly (in 1 h) but seldom in better yield than the conventional
approach (4−21 days). The coumarins 7a−e are also known.5
Hydrochloric Acid-Catalyzed Reactions of the Baylis−Hill-
man Adducts 5a−e. The general procedure is illustrated by the
following example.
CONCLUSIONS
■
1H NMR-based kinetic studies of the hydrochloric acid-catalyzed
reaction of the salicylaldehyde-derived Baylis−Hillman adduct
5a have permitted evaluation of the kinetic parameters for the
pseudo-first-order dissociation of intermediates 10a and 13a
leading to the major and minor products 6a and 7a, respectively.
Independent pathways have been identified for the formation of
these products at low temperatures (276.2−293.4 K), and
detailed theoretical studies have provided important insights into
the mechanistic implications and permitted rationalization of
their latent complexity. Theoretical studies of the acetic acid-
catalyzed thermal cyclization of 6a to give 7a have indicated the
possible involvement of a hydrogen-bonded quinone methide
transition-state complex in the critical cyclization stepan
Acetyl chloride (1.2 mL) was added dropwise to dry methanol (1.0
mL) cooled in an ice bath (CAUTION! exothermic reaction). The
mixture was left to stir for 30 min, and tert-butyl 3-hydroxy-3-(2-
hydroxyphenyl)-2-methylenepropanoate (5a) (0.50 g, 2.0 mmol)
dissolved in dry methanol (2 mL) was then added. The mixture was
left to stir overnight. Water (20 mL) was added to the crude mixture,
which was then extracted with CHCl3 (2 × 20 mL) and concentrated in
vacuo. Fractional recrystallization from chloroform afforded two
products, the major product (Z)-2-(chloromethyl)-3-(2-hydroxyphenyl)-
2-propenoic acid (6a) [pink solid (0.35 g, 83%), mp 129−131 °C
(Found: C, 56.22; H, 4.40%. C10H9ClO3 requires C, 56.49; H, 4.27%);
νmax/cm−1 3453 (OH), 1666 (CO); δH (400 MHz, DMSO-d6) 4.47
(2H, s), 6.90−6.96 (2H, m), 7.28 (1H, t, J = 7.7 Hz), 7.49 (1H, m), 7.96
(1H, s), 10.22 (1H, br s); δC (100 MHz, DMSO-d6) 41.1, 116.7, 120.1,
116
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
Figure 9. Mass-balanced free energy profile for the formation of the minor product 7a via O-acyl cleavage (path B), showing ball-and-stick structures for
the transition states TS IV−TS VI, colored according to atom type (H, white; C, gray; O, red; Cl, green).
Figure 10. Superimposed mass-balanced free energy profiles for the formation of the major product 6a via O-alkyl cleavage (path A) and the minor
product 7a via O-acyl cleavage (path B).
121.8, 128.2, 130.0, 132.3, 139.5, 157.4, 168.2] and the minor product 3-
(chloromethyl)coumarin (7a) [pale-yellow solid (0.05 g, 13%)].
Yields and analytical data for the remaining products are as follows:
(Z)-3-(5-Bromo-2-hydroxyphenyl)-2-(chloromethyl)-2-propenoic acid
(6b) [pale-yellow solid (0.37 g, 84%), mp 146−148 °C (Found: C,
41.29; H, 2.88%. C10H8BrClO3 requires C, 41.20; H, 2.77%); νmax/cm−1
117
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
Figure 11. Mass-balanced free energy profile for the Z−E isomerization of 2-(chloromethyl)cinnamic acid derivative 6a and subsequent cyclization to
give 3-(chloromethyl)coumarin 7a (path C), showing ball-and-stick structures for the transition states TS IV−TS VI, colored according to atom type
(H, white; C, gray; O, red; Cl, green).
Figure 12. (a) Mechanistic depiction of the role of acetic acid as an amphoteric catalyst in the cyclization of (Z)-3-(chloromethyl)-2-(hydroxyphenyl)-2-
propenoic acid (15a). (b) Putative structures involved in a suggested enzyme-catalyzed E−Z isomerization of o-coumaric acid 16 and the subsequent
lactonization of the Z isomer 17 to form coumarin.
3421 (OH), 1669 (CO); δH (400 MHz, methanol-d4) 4.43 (2H, s),
6.81 (1H, d, J = 8.7 Hz), 7.36 (1H, d, J = 8.7 Hz), 7.71 (1H, s), 7.97 (1H,
s); δC (100 MHz, DMSO-d6) 40.7, 111.0, 118.7, 124.0, 129.5, 132.1,
134.6, 138.0, 156.6, 167.8] and 6-bromo-3-(chloromethyl)coumarin (7b)
(14% estimated by 1H NMR analysis).
(Z)-2-(Chloromethyl)-3-(2-hydroxy-3-methoxyphenyl)-2-propenoic
acid (6c) [lilac solid (0.39 g, 90%), mp 137−139 °C (Found: C, 54.63;
H, 4.80%. C11H11ClO4 requires C, 54.45; H, 4.57%); νmax/cm−1 3499
(OH), 1671 (CO); δH (400 MHz, methanol-d4) 3.88 (3H, s), 4.47
(2H, s), 6.88 (1H, t, J = 7.9 Hz), 7.00 (1H, d, J = 7.9 Hz), 7.20 (1H, d, J =
7.9 Hz), 8.10 (1H, s); δC (100 MHz, methanol-d4) 40.6, 56.5, 113.8,
120.3, 122.2, 122.6, 129.2, 140.5, 147.2, 149.1, 169.7] and 3-
(chloromethyl)-8-methoxycoumarin (7c) [brown solid (0.04 g, 10%)].
(Z)-2-(Chloromethyl)-3-(3-ethoxy-2-hydroxyphenyl)-2-propenoic acid
(6d) [purple solid (0.37 g, 85%), mp 128−130 °C (Found: C, 56.27;
H, 5.10%. C12H13ClO4 requires C, 56.15; H, 5.10%); νmax/cm−1 3489
(OH), 1674 (CO); δH (400 MHz, CDCl3) 1.47 (3H, t, J = 6.9 Hz),
4.15 (2H, q, J = 6.9 Hz), 4.50 (2H, s), 6.07 (1H, br s), 6.92 (2H, m), 7.28
(1H, m), 8.25 (1H, s); δC (100 MHz, CDCl3) 14.8, 39.3, 64.8, 113.1,
119.9, 120.4, 121.4, 127.3, 140.7, 145.1, 145.8, 171.3] and 3-
(chloromethyl)-8-ethoxycoumarin (7d) (10% estimated by 1H NMR
analysis).
118
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
(Z)-3-(5-Chloro-2-hydroxyphenyl)-2-(chloromethyl)-2-propenoic acid
(6e) [pale-yellow solid (0.39 g, 89%), mp 126−128 °C (Found: C,
48.48; H, 3.48%. C10H8Cl2O3 requires C, 48.61; H, 3.26%); νmax/cm−1
3408 (OH), 1666 (CO); δH (400 MHz, CDCl3) 4.42 (2H, s), 6.80
(1H, d, J = 8.7 Hz), 7.21 (1H, d, J = 8.7 Hz), 7.55 (1H, s), 8.00 (1H, s);
δC (100 MHz, methanol-d4) 40.1, 117.9, 124.3, 125.2, 130.0, 130.3,
131.9, 139.3, 156.5, 169.3 (CO)] and 6-chloro-3-(chloromethyl)-
coumarin (7e) (10% estimated by 1H NMR analysis).
ACKNOWLEDGMENTS
■
The authors thank Rhodes University for a bursary (to T.O.O.).
This work is based upon research supported by the National
Research Foundation (NRF; South Africa: IFR2011032500034)
and Rhodes University. T.O.O. gratefully acknowledges Obafemi
Awolowo University (Ile-Ife, Nigeria) for study leave. M.R.C.
thanks the University of Cape Town and the NRF (South Africa:
IFR201132500034) for research support. Any opinion, findings,
and conclusions or recommendations expressed in this material
are those of the authors, and therefore, the NRF does not accept
any liability in regard thereto.
X-ray Analysis of 2-(Chloromethyl)cinnamic Acid Derivative
6a. Crystal data for 6a: C10H9ClO3, M = 212.62, 0.25 mm × 0.13 mm ×
0.09 mm, triclinic, space group P1 (No. 2), a = 5.1208(4) Å, b =
̅
7.8498(8) Å, c = 11.9863(12) Å, α = 99.256(4)°, β = 91.114(7)°, γ =
100.217(7)°, V = 467.43(8) Å3, Z = 2, Dc = 1.511 g cm−3, F000 = 220, Mo
Kα radiation, λ = 0.71073 Å, T = 173(2) K, 2θmax = 54.9°, 25 716
reflections collected, 2115 unique (Rint = 0.0427). Final GOF = 1.044, R1
= 0.0765, wR2 = 0.2063, R indices based on 1552 reflections with I >
2σ(I), (refinement on F2), 129 parameters, 0 restraints. Lorentz
polarization and absorption corrections applied, μ = 0.383 mm−1,
CCDC 1016110.
REFERENCES
■
(1) (a) Baylis, A. B.; Hillman, M. E. D. German Patent 2155113, 1972.
(b) Takizawa, S.; Remond, E.; Arteaga, F. A.; Yoshida, Y.; Sridharan, V.;
́
́
Bayardon, J.; Juge, S.; Sasai, H. Chem. Commun. 2013, 49, 8392−8394.
(c) Peng, C.; Joy, A. Macromolecules 2014, 47, 1258−1268. (d) Verma,
P.; Verma, P.; Sunoj, R. B. Org. Biomol. Chem. 2014, 12, 2176−2179.
(e) Basavaiah, D.; Veeraraghavaiah, G.; Badsara, S. S. Org. Biomol. Chem.
2014, 12, 1551−1555. (f) Yin, Y.; Sun, G.; Zhang, H.; Zhou, H.; Wu, F.
Chin. J. Chem. 2014, 32, 365−369. (g) Bharadwaj, K. C. RSC Adv. 2015,
5, 75923. (h) Xie, P.; Huang, Y. Org. Biomol. Chem. 2015, 13, 8578.
(2) (a) Morita, K.; Suzuki, Z.; Hirose, H. Bull. Chem. Soc. Jpn. 1968, 41,
2815. (b) Morita, K. Japanese Patent 6803364, 1968.
Kinetic Studies. Kinetic experiments were conducted on a Bruker
Biospin 600 MHz NMR spectrometer. Temperature calibration of the
spectrometer was carried out between 273 and 333 K, and the
temperatures reported for each kinetic run are the corrected values. For
each kinetic run, acetyl chloride (0.1 mL) was added to cooled CD3OD
(0.4 mL) in a graduated NMR tube, which was then sealed with a
1
septum, and an initial H NMR spectrum was obtained. A solution of
compound 5a (20 mg, 80 μmol) in CD3OD (0.1 mL) was then added,
the total volume of reaction mixture noted, and the tube replaced in the
NMR probe. 1H NMR spectra were recorded automatically with a delay
of 40 s between acquisitions. Experiments were repeated at six different
temperatures between 273 and 295 K.
(3) (a) Basavaiah, D.; Rao, K. V.; Reddy, R. J. Chem. Soc. Rev. 2007, 36,
1581−1588. (b) Singh, V.; Batra, S. Tetrahedron 2008, 64, 4511−4574.
(c) Basavaiah, D.; Reddy, B. S.; Badsara, S. S. Chem. Rev. 2010, 110,
5447−5674. (d) Wei, Y.; Shi, M. Chem. Rev. 2013, 113, 6659−6690.
(e) Ciganek, E. Org. React. 1997, 51, 201−350.
Computational Studies. All of the density functional theory
calculations were performed with the B3LYP hybrid density functional
with the 6.31G(d) basis set using the Gaussian 03 program17 running on
an Intel/Linux cluster; Gaussview 4.126 and Molden27 were used for
visualization. Stationary-state structures were characterized by vibra-
tional analysis. Substrates, products, and intermediates were charac-
terized by the absence of imaginary frequencies and transition-state
complexes by the presence of a single imaginary frequency. Each
transition-state structure was confirmed by inspection of the vibration
corresponding to the imaginary frequency, calculation of the
corresponding IRC path, or in some cases energy minimization of
structures close to the transition state. Table 5 summarizes the
calculated free energies and, where appropriate, vibrational frequencies
for the stationary states involved in paths A, B, and C.
(4) (a) Familoni, O. B.; Kaye, P. T.; Klaas, P. J. Chem. Commun. 1998,
2563−2564. (b) Familoni, O. B.; Klaas, P. J.; Lobb, K. A.; Pakade, V. E.;
Kaye, P. T. Org. Biomol. Chem. 2006, 4, 3960−3965.
(5) Kaye, P. T.; Musa, M. A.; Nocanda, X. W.; Robinson, R. S. Org.
Biomol. Chem. 2003, 1, 1133−1138.
(6) (a) Kaye, P. T.; Nocanda, X. W. Synthesis 2001, 2389−2392.
(b) Kaye, P. T.; Nocanda, X. W. J. Chem. Soc., Perkin. Trans. 1 2002,
1318−1323.
(7) Olomola, T. O.; Klein, R.; Lobb, K. A.; Sayed, Y.; Kaye, P. T.
Tetrahedron Lett. 2010, 51, 6325−6328.
(8) Olomola, T. O.; Klein, R.; Mautsa, N.; Sayed, Y.; Kaye, P. T. Bioorg.
Med. Chem. 2013, 21, 1964−1971.
(9) Didierjean, J.; Isel, C.; Querre, F.; Mouscadet, J.-F.; Aubertin, A.-
M.; Valnot, J.-Y.; Piettre, S. R.; Marquet, R. Antimicrob. Agents
Chemother. 2005, 49, 4884−4894.
ASSOCIATED CONTENT
(10) Wang, Z.; Bennett, E. M.; Wilson, D. J.; Salomon, C.; Vince, R. J.
Med. Chem. 2007, 50, 3416−3419.
■
S
* Supporting Information
(11) (a) Dayam, R.; Gundla, R.; Al-Mawsawi, L. Q.; Neamati, N. Med.
Res. Rev. 2008, 28, 118−154. (b) Bodiwala, H. S.; Sabde, S.; Gupta, P.;
Mukherjee, R.; Kumar, R.; Garg, P.; Bhutani, K. K.; Mitra, D.; Singh, I. P.
Bioorg. Med. Chem. 2011, 19 (3), 1256−1263.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.5b02372.
Copies of 1H and 13C NMR spectra for all new compounds
(12) Idahosa, K. C.; Lee, Y.-C.; Nyoni, D.; Kaye, P. T.; Caira, M. R.
Tetrahedron Lett. 2011, 52, 2972−2976.
and X-ray crystal structure of compound 6a (PDF)
Cartesian coordinates of all computed structures (TXT)
(13) (a) Hill, J. S.; Isaacs, N. S. J. Phys. Org. Chem. 1990, 3, 285−290.
(b) Bode, M. L.; Kaye, P. T. Tetrahedron Lett. 1991, 32, 5611−5614.
(c) Price, K. E.; Broadwater, S. J.; Jung, H. M.; McQuade, D. T. Org. Lett.
2005, 7, 147−150. (d) Price, K. E.; Broadwater, S. J.; Walker, D. J.;
McQuade, D. T. J. Org. Chem. 2005, 70, 3980−3987. (e) Robiette, R.;
Aggarwal, V. K.; Harvey, J. N. J. Am. Chem. Soc. 2007, 129, 15513−
15525.
X-ray crystallographic data for compound 6a (CIF)
The crystallographic data for compound 6a have also been
deposited with the Cambridge Crystallographic Data Centre
(CCDC 1016110).
(14) Olomola, T. O.; Klein, R.; Kaye, P. T. Synth. Commun. 2012, 42,
251−257.
(15) Tiekenk, E. R. T. In Organic Crystal Engineering: Frontiers in
Crystal Engineering; Tiekenk, E. R. T., Vittal, J., Zaworotko, M., Eds.;
John Wiley & Sons: Chichester, U.K., 2010; Chapter 6, pp 191−192.
(16) Bernstein, J.; Davis, R. E.; Shimoni, L.; Chang, N. L. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1555−1573.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: P.Kaye@ru.ac.za.
Notes
The authors declare no competing financial interest.
119
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120

The Journal of Organic Chemistry
Article
(17) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K.
N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004.
(18) Khene, S.; Lobb, K.; Nyokong, T. Electrochim. Acta 2010, 56 (2),
706−716.
(19) Perkin, W. H. J. Chem. Soc. 1868, 21 (53), 181−186.
(20) Ranu, B. C.; Jana, R. Eur. J. Org. Chem. 2006, 2006, 3767−3770.
(21) Patre, R. E.; Parameswaran, P. S.; Tilve, S. G. ARKIVOC 2011,
2011 (ix), 68−76.
(22) Jeon, J.-H.; Yang, D.-M.; Jun, J.-G. Bull. Korean Chem. Soc. 2011,
32, 65−70.
(23) Stoker, J. R.; Bellis, D. M. J. Biol. Chem. 1962, 237 (7), 2303−
2305.
(24) Haskins, F. A.; Williams, L. G.; Gorz, H. J. Plant Physiol. 1964, 39,
777−781.
(25) Bayoumi, S. A. L.; Rowan, M. G.; Blagbrough, I. S.; Beeching, J. R.
Phytochemistry 2008, 69, 2928−2936.
(26) Dennington, R.; Keith, T.; Millam, J. GaussView, version 4.1;
Semichem Inc.: Shawnee Mission, KS, 2009.
(27) Schaftenaar, G.; Noordik, J. H. J. Comput.-Aided Mol. Des. 2000,
14, 123−134.
120
DOI: 10.1021/acs.joc.5b02372
J. Org. Chem. 2016, 81, 109−120